Inductive Unsupervised Domain Adaptation for Few-Shot Classification via Clustering

TL;DR

This paper introduces DaFeC, an inductive clustering-based framework that enhances few-shot domain adaptation without requiring test data, achieving significant accuracy improvements on the FewRel 2.0 dataset.

Contribution

Proposes an inductive framework for few-shot domain adaptation using clustering, pseudo-labeling, and feature enhancement techniques, avoiding reliance on test data.

Findings

Outperforms previous methods with up to 11.62% accuracy gain

Effective pseudo-labeling via clustering and similarity minimization

Improved feature learning with adversarial alignment

Abstract

Few-shot classification tends to struggle when it needs to adapt to diverse domains. Due to the non-overlapping label space between domains, the performance of conventional domain adaptation is limited. Previous work tackles the problem in a transductive manner, by assuming access to the full set of test data, which is too restrictive for many real-world applications. In this paper, we set out to tackle this issue by introducing a inductive framework, DaFeC, to improve Domain adaptation performance for Few-shot classification via Clustering. We first build a representation extractor to derive features for unlabeled data from the target domain (no test data is necessary) and then group them with a cluster miner. The generated pseudo-labeled data and the labeled source-domain data are used as supervision to update the parameters of the few-shot classifier. In order to derive high-quality pseudo labels, we propose a Clustering Promotion Mechanism, to learn better features for the target domain via Similarity Entropy Minimization and Adversarial Distribution Alignment, which are combined with a Cosine Annealing Strategy. Experiments are performed on the FewRel 2.0 dataset. Our approach outperforms previous work with absolute gains (in classification accuracy) of 4.95%, 9.55%, 3.99% and 11.62%, respectively, under four few-shot settings.